feat: improve resources_monitoring for OOM & OOD monitoring (#16241)

As requested for [this
issue](https://github.com/coder/internal/issues/245) we need to have a
new resource `resources_monitoring` in the agent.

It needs to be parsed from the provisioner and inserted into a new db
table.
This commit is contained in:
Vincent Vielle
2025-02-04 18:45:33 +01:00
committed by GitHub
parent 8c265018c4
commit 7cbd77fd94
76 changed files with 3170 additions and 1041 deletions
+32
View File
@@ -1391,6 +1391,14 @@ func (q *querier) FavoriteWorkspace(ctx context.Context, id uuid.UUID) error {
return update(q.log, q.auth, fetch, q.db.FavoriteWorkspace)(ctx, id)
}
func (q *querier) FetchMemoryResourceMonitorsByAgentID(ctx context.Context, agentID uuid.UUID) (database.WorkspaceAgentMemoryResourceMonitor, error) {
if err := q.authorizeContext(ctx, policy.ActionRead, rbac.ResourceWorkspaceAgentResourceMonitor); err != nil {
return database.WorkspaceAgentMemoryResourceMonitor{}, err
}
return q.db.FetchMemoryResourceMonitorsByAgentID(ctx, agentID)
}
func (q *querier) FetchNewMessageMetadata(ctx context.Context, arg database.FetchNewMessageMetadataParams) (database.FetchNewMessageMetadataRow, error) {
if err := q.authorizeContext(ctx, policy.ActionRead, rbac.ResourceNotificationMessage); err != nil {
return database.FetchNewMessageMetadataRow{}, err
@@ -1398,6 +1406,14 @@ func (q *querier) FetchNewMessageMetadata(ctx context.Context, arg database.Fetc
return q.db.FetchNewMessageMetadata(ctx, arg)
}
func (q *querier) FetchVolumesResourceMonitorsByAgentID(ctx context.Context, agentID uuid.UUID) ([]database.WorkspaceAgentVolumeResourceMonitor, error) {
if err := q.authorizeContext(ctx, policy.ActionRead, rbac.ResourceWorkspaceAgentResourceMonitor); err != nil {
return nil, err
}
return q.db.FetchVolumesResourceMonitorsByAgentID(ctx, agentID)
}
func (q *querier) GetAPIKeyByID(ctx context.Context, id string) (database.APIKey, error) {
return fetch(q.log, q.auth, q.db.GetAPIKeyByID)(ctx, id)
}
@@ -3003,6 +3019,14 @@ func (q *querier) InsertLicense(ctx context.Context, arg database.InsertLicenseP
return q.db.InsertLicense(ctx, arg)
}
func (q *querier) InsertMemoryResourceMonitor(ctx context.Context, arg database.InsertMemoryResourceMonitorParams) (database.WorkspaceAgentMemoryResourceMonitor, error) {
if err := q.authorizeContext(ctx, policy.ActionCreate, rbac.ResourceWorkspaceAgentResourceMonitor); err != nil {
return database.WorkspaceAgentMemoryResourceMonitor{}, err
}
return q.db.InsertMemoryResourceMonitor(ctx, arg)
}
func (q *querier) InsertMissingGroups(ctx context.Context, arg database.InsertMissingGroupsParams) ([]database.Group, error) {
if err := q.authorizeContext(ctx, policy.ActionCreate, rbac.ResourceSystem); err != nil {
return nil, err
@@ -3195,6 +3219,14 @@ func (q *querier) InsertUserLink(ctx context.Context, arg database.InsertUserLin
return q.db.InsertUserLink(ctx, arg)
}
func (q *querier) InsertVolumeResourceMonitor(ctx context.Context, arg database.InsertVolumeResourceMonitorParams) (database.WorkspaceAgentVolumeResourceMonitor, error) {
if err := q.authorizeContext(ctx, policy.ActionCreate, rbac.ResourceWorkspaceAgentResourceMonitor); err != nil {
return database.WorkspaceAgentVolumeResourceMonitor{}, err
}
return q.db.InsertVolumeResourceMonitor(ctx, arg)
}
func (q *querier) InsertWorkspace(ctx context.Context, arg database.InsertWorkspaceParams) (database.WorkspaceTable, error) {
obj := rbac.ResourceWorkspace.WithOwner(arg.OwnerID.String()).InOrg(arg.OrganizationID)
tpl, err := q.GetTemplateByID(ctx, arg.TemplateID)
+93
View File
@@ -4562,3 +4562,96 @@ func (s *MethodTestSuite) TestOAuth2ProviderAppTokens() {
}).Asserts(rbac.ResourceOauth2AppCodeToken.WithOwner(user.ID.String()), policy.ActionDelete)
}))
}
func (s *MethodTestSuite) TestResourcesMonitor() {
s.Run("InsertMemoryResourceMonitor", s.Subtest(func(db database.Store, check *expects) {
dbtestutil.DisableForeignKeysAndTriggers(s.T(), db)
check.Args(database.InsertMemoryResourceMonitorParams{}).Asserts(rbac.ResourceWorkspaceAgentResourceMonitor, policy.ActionCreate)
}))
s.Run("InsertVolumeResourceMonitor", s.Subtest(func(db database.Store, check *expects) {
dbtestutil.DisableForeignKeysAndTriggers(s.T(), db)
check.Args(database.InsertVolumeResourceMonitorParams{}).Asserts(rbac.ResourceWorkspaceAgentResourceMonitor, policy.ActionCreate)
}))
s.Run("FetchMemoryResourceMonitorsByAgentID", s.Subtest(func(db database.Store, check *expects) {
u := dbgen.User(s.T(), db, database.User{})
o := dbgen.Organization(s.T(), db, database.Organization{})
tpl := dbgen.Template(s.T(), db, database.Template{
OrganizationID: o.ID,
CreatedBy: u.ID,
})
tv := dbgen.TemplateVersion(s.T(), db, database.TemplateVersion{
TemplateID: uuid.NullUUID{UUID: tpl.ID, Valid: true},
OrganizationID: o.ID,
CreatedBy: u.ID,
})
w := dbgen.Workspace(s.T(), db, database.WorkspaceTable{
TemplateID: tpl.ID,
OrganizationID: o.ID,
OwnerID: u.ID,
})
j := dbgen.ProvisionerJob(s.T(), db, nil, database.ProvisionerJob{
Type: database.ProvisionerJobTypeWorkspaceBuild,
})
b := dbgen.WorkspaceBuild(s.T(), db, database.WorkspaceBuild{
JobID: j.ID,
WorkspaceID: w.ID,
TemplateVersionID: tv.ID,
})
res := dbgen.WorkspaceResource(s.T(), db, database.WorkspaceResource{JobID: b.JobID})
agt := dbgen.WorkspaceAgent(s.T(), db, database.WorkspaceAgent{ResourceID: res.ID})
dbgen.WorkspaceAgentMemoryResourceMonitor(s.T(), db, database.WorkspaceAgentMemoryResourceMonitor{
AgentID: agt.ID,
Enabled: true,
Threshold: 80,
CreatedAt: dbtime.Now(),
})
monitor, err := db.FetchMemoryResourceMonitorsByAgentID(context.Background(), agt.ID)
require.NoError(s.T(), err)
check.Args(agt.ID).Asserts(rbac.ResourceWorkspaceAgentResourceMonitor, policy.ActionRead).Returns(monitor)
}))
s.Run("FetchVolumesResourceMonitorsByAgentID", s.Subtest(func(db database.Store, check *expects) {
u := dbgen.User(s.T(), db, database.User{})
o := dbgen.Organization(s.T(), db, database.Organization{})
tpl := dbgen.Template(s.T(), db, database.Template{
OrganizationID: o.ID,
CreatedBy: u.ID,
})
tv := dbgen.TemplateVersion(s.T(), db, database.TemplateVersion{
TemplateID: uuid.NullUUID{UUID: tpl.ID, Valid: true},
OrganizationID: o.ID,
CreatedBy: u.ID,
})
w := dbgen.Workspace(s.T(), db, database.WorkspaceTable{
TemplateID: tpl.ID,
OrganizationID: o.ID,
OwnerID: u.ID,
})
j := dbgen.ProvisionerJob(s.T(), db, nil, database.ProvisionerJob{
Type: database.ProvisionerJobTypeWorkspaceBuild,
})
b := dbgen.WorkspaceBuild(s.T(), db, database.WorkspaceBuild{
JobID: j.ID,
WorkspaceID: w.ID,
TemplateVersionID: tv.ID,
})
res := dbgen.WorkspaceResource(s.T(), db, database.WorkspaceResource{JobID: b.JobID})
agt := dbgen.WorkspaceAgent(s.T(), db, database.WorkspaceAgent{ResourceID: res.ID})
dbgen.WorkspaceAgentVolumeResourceMonitor(s.T(), db, database.WorkspaceAgentVolumeResourceMonitor{
AgentID: agt.ID,
Path: "/var/lib",
Enabled: true,
Threshold: 80,
CreatedAt: dbtime.Now(),
})
monitors, err := db.FetchVolumesResourceMonitorsByAgentID(context.Background(), agt.ID)
require.NoError(s.T(), err)
check.Args(agt.ID).Asserts(rbac.ResourceWorkspaceAgentResourceMonitor, policy.ActionRead).Returns(monitors)
}))
}